Following our effort in monitoring U.S. local open data policies, we decided to examine municipalities outside the country to find out how their policies are measuring up to Sunlight’s Open Data Policy Guidelines — starting with Canada, which signed the G8 Open Data Charter more than a year ago.
After a long delay, most G7 countries eventually released their Open Data Action Plans to realize their commitments to five core principles, including making data open by default. While multilateral agreements always catch the eyeballs of media and the public, it is as important for us to follow the development of local open data initiatives. Cities are the bases for political interactions between people and government. Local authorities are responsible for releasing information relevant to our daily lives, such as approval notices for changing land use near your home, or a council’s decision to remove the bus line you ride every day.
Although some cities, such as Paris, have started incremental steps to incorporate open data principles into local policies, many others are still missing. In contrast, many Canadian municipalities have taken the lead to formulate open data policies at the city level, some even before the signing of the G8 Charter in 2013. Therefore, we reviewed 13 municipal open data policies (or de facto policies) in Canada, as a preliminary case study.
Even though Canada has taken the most positive steps among the G7, there is still room for improvement for their municipal policies regarding what data should be public, how to make data public and how to implement policy.
“Open by default” is rarely mentioned
The first principle of the G8 Charter is that government data should be “open by default.” This principle means that “all government data” will be published openly unless specified with “legitimate reasons,” and it applies to “national, federal, local, or international government.” However, among the Canadian municipal policies that we found, only the City of Kitchener mentioned the exact phrase of “open by default” in their policy commitments. Five other cities have a close fit, such as “making the disclosure of government-held information an automatic process wherever possible” (Toronto) and sharing “the greatest amount of data possible”(Hamilton). However, the City of London and Brampton take the opposite approach. The former sat up a working group to “develop criteria to guide decision-making around what data sets may be released,” and the latter would only open data “that the City has approved for.” We believe such an approach falls short of proactive disclosure in our guidelines and has the potential to hinder the openness of data.
Formats of data are not specified
Formats of the open data are important. Sunlight goes a long way to explain and show how release of raw data, like Application Program Interfaces (APIs), metadata, bulk data and open source solutions are necessary to improve access to information and develop useful tools. However, almost none of these are required to be released by Canadian municipal open data policies. For instance, with the exception of Toronto, no city authorities are mandated to release APIs and bulk data, according to the policies we found. Although some cities have provided APIs, such as Montréal, and unique identifiers for some datasets in practice, there is an increased potential for inconsistencies among individual departments and agencies without a clear policy mandate.
Data collection, inventories and other trends
In addition, some other important guidelines are barely mentioned. Although most policies require authorities to publish a category of data that they release, almost none of them (except the Municipality of York) mandate for publishing data inventories or a “comprehensive list of all information holdings.” None of these 13 policies require officials to “optimize methods of data collection,” such as using electronic-filing to facilitate data release. Only Vancouver and Montréal are able to mention that open data policies are also applicable to third-party data producers like contractors and consultants. Meanwhile, Brampton restricted its policy only to “datasets generated by the City and upon which the City relies.” It may, as a result, exclude most contractors or quasi-governmental agencies from the realm of scrutiny, while they often produce important documents such as procurement contracts or major infrastructure project updates.
Although this research is by no means exhaustive, it serves an illustrative purpose of how local policies can be formulated differently, even within the same country or the same province. It is important for us to keep track of open data policies at the local level, as they have tremendous and direct influences on how government releases data to citizens every day. After that, the question we should ask is about compliance. Currently, Canada has about 50 municipal open data portals available online, but there is still a long way to go to fully realize the G8 Open Data Charter to “national, federal [and] local government.”
The following is our full list comparison on Canadian municipal open data policies: